1100-line Perl emulator for BBN-Lisp runs original Eliza program(my.umbc.edu)
my.umbc.edu
1100-line Perl emulator for BBN-Lisp runs original Eliza program
http://my.umbc.edu/news/48895
8 comments
I'd really like to see PARRY (http://en.wikipedia.org/wiki/PARRY) run on a modern system: http://www.cs.cmu.edu/afs/cs/project/ai-repository/ai/areas/...
If you're interested in chatterbot, check out a short piece I wrote the state of the art - https://medium.com/@rob_ellis/creating-a-chat-bot-42861e6a2a... and a new framework superscriptjs.com
You could easily port PARRY to work on SuperScript.
You could easily port PARRY to work on SuperScript.
The title had me confused for a bit. It's a Lisp emulator written in Perl, not a Perl emulator in Lisp. I was actually hoping it would be the latter, simply because that would be impressive.
GitHub repo: https://github.com/jeffshrager/elizagen
On the one hand, this is really cool. On the other, I'm always sad to see projects like this which seem to follow Perl 4 Best Practices type design. By this, I mean ...
* No strict or warnings
* Calling subs with "&foo" syntax
* Globals, globals everywhere
It's cool that Perl just lets you whip things up with so little ceremony, but it's always nice when people write nice code, in any language.
* No strict or warnings
* Calling subs with "&foo" syntax
* Globals, globals everywhere
It's cool that Perl just lets you whip things up with so little ceremony, but it's always nice when people write nice code, in any language.
I always love it when someone does something awesome in Perl, but always a little scared to see the code :-(
A sad thing here is that the code looks like it mostly would run under strict and warnings, and it'd be pretty easy to simplify the sub definitions at the start
A sad thing here is that the code looks like it mostly would run under strict and warnings, and it'd be pretty easy to simplify the sub definitions at the start
Strict is not that big of a deal and can be easily enabled later. Seriously, those issues with the code are really tiny and very unimportant, especially for this project. But I would definitely choose this kind of code over anything with Moo* for example.
It's sad that people don't know or forget what makes a code easier to understand and how important it is.
It's sad that people don't know or forget what makes a code easier to understand and how important it is.
Why is someone interested in old Lisp, if their best attempt to emulate it inspires them to code in Perl.
It's like uncovering the manuscript of some old Beethoven score, and you and your friends choose to squeak it out with your armpits, instead of pulling together a string quartet.
It's like uncovering the manuscript of some old Beethoven score, and you and your friends choose to squeak it out with your armpits, instead of pulling together a string quartet.